ClawHub
Back to skill

Security audit

failure-forensics

Security checks across malware telemetry and agentic risk

Overview

This skill performs disclosed, local post-mortem analysis of agent failures, but users should treat logs and saved reports as potentially sensitive.

Install only if you want agents to analyze failure logs and create persistent post-mortem artifacts. Before saving or sharing reports, review and redact tokens, credentials, personal data, internal paths, prompts, and sensitive command output; store reports only in locations with appropriate access controls.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
93% confidence
Finding
The skill instructs reading logs, generating reports, saving artifacts to persistent storage, and references network-related failure analysis, but it declares no permissions. That mismatch is dangerous because it can cause an agent or reviewer to underestimate the skill's access needs, allowing log ingestion, report persistence, or network-touching behavior to occur without explicit scrutiny or least-privilege controls.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The README explicitly promotes parsing tool-call logs and generating/saving post-mortem reports, but it does not warn that such logs often contain sensitive material such as command arguments, file paths, tokens, environment details, or error output. In an agent-forensics context, this omission is more dangerous because failure logs are likely to capture exactly the sensitive operational context surrounding a failed task, increasing the risk of accidental retention, disclosure, or propagation into generated reports.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill explicitly tells the agent to save post-mortem reports and lessons to durable storage, but it does not warn that logs and reports may contain credentials, tokens, internal paths, prompts, or other sensitive operational data. This creates a real risk of inadvertent persistence and later disclosure of secrets or confidential incident details, especially because forensic artifacts often aggregate exactly the most sensitive context from a failed run.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.